Project 3 Solution

For this assignment, you will use C++ to implement two different algorithms for multiplying very large integers. The purpose is to give you some additional experience translating a non-trivial algorithm from pseudocode to a working implementation.

Your Task    You should write a C++ program that does these things:

    1. Read two digit strings a and b, separated by a * character, from standard input. The numbers a and b may be very large, potentially thousands of digits each. Either or both of the numbers may be zero.

    2. Multiply a and b using some  (n2) time brute force algorithm. The text book mentions one such algorithm on page 187 as the “pencil and paper” algorithm. Any method typically taught to small children will work.

    3. Output the text “B: ”, followed by the answer from Step (2).

    4. Multiply a and b again using the divide-and-conquer Karatsuba multiplication algorithm.

    5. Output the text “K: ”, followed by the answer from Step (4).

That’s it. Just multiply the two numbers using both algorithms.

Sample Input 1
1234*5678

Sample Output 1

    B: 7006652

    K: 7006652


Sample Input 2
123456789123456789*987654321987654321

Sample Output 2

    B: 121932631356500531347203169112635269

    K: 121932631356500531347203169112635269


Sample Input 3
480206450526460879603316532635846*21789169871971286

Sample Output 3

    B: 10463299924137431288618787106491550779513606317956

    K: 10463299924137431288618787106491550779513606317956



csce350    Project 3    1 of 2

Hints    A few possibly helpful comments:

The most common error for this kind of program is to try to use an int variable to store the input, the output, or values computed along the way. Remember that most C++ compilers use 32 bits of memory to store int variables. This means that for numbers larger than about 232 = 4; 294; 967; 296, things will begin to fail because of overflow. These kinds of overflow problems will not generate any error messages. Because you want a program that works for numbers much larger than 232, you must store numbers as vectors of digits at every step along the way.

You may use (and indeed, probably should) use the STL vector class for the “array” that holds the digits in each number you manipulate. This simplifies some aspects of the problem because it is much easier to change the size of a vector than of an C++ array.

You will want to think carefully about the order in which the digits are stored in your vector of digits. Which digit goes in position 0? Which digit goes in position i? There are two choices for how to organize things, both of which can work just fine, but each has advantages and disadvantages.

When you divide the two numbers a and b into a0, a1, b0 and b1, be careful to use the same value of m for both a and b.

You may generate additional debugging output for each algorithm if you like, as long as the two lines required in Steps (3) and (5) appear eventually. Such extra outputs may be useful in awarding partial credit if your solution is not fully correct.

Watch out for the case that a and b have different numbers of digits. The easiest way to handle this is to “pad” the shorter number with leading 0’s to equalize the lengths.

For calibration purposes, my solution has 266 lines of code, including comments.

What to Submit You should submit, using the department’s dropbox website, a single C++ source file named containing all of the code for your program. I will compile this program us-ing this command line:

g++ -Wall -std=c++11 yourfile.cpp




















csce350    Project 3    2 of 2